摘要
针对Apriori算法存在的问题,提出了一种基于待与项集的频繁项集挖掘算法,从两方面考虑对算法效率进行改进:通过对项编码来减少扫描数据库次数;提出了一个新的概念—待与项集,通过从待与项集中删除项来减少候选项集的数量。实例分析表明,该方法仅需扫描一次数据库,而且具有搜索速度快、节省内存空间等优点。该算法同样适用于处理超大型事务数据库。
The problem of Apriori algorithm is discussed, and an algorithm of mining frequent itemsets based on pending items is proposed. The mining efficiency is improved from two aspects. One is to decrease the scanning times over database by means of coding for every item. The other is to reduce the candidate itemsets by deleting items from pending items, which is advanced as a new concept. Analysis of examples proves that it can not only scan the database once, but also has the virtues in high speed, less memory cost. This algorithm is also fit for dealing with super transaction database.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第1期129-131,共3页
Computer Engineering and Design
